ABSTRACT The multi-disciplinary accelerator based facilities at iThemba LABS are used intensively for nuclear physics experiments, radiotherapy and the production of radioisotopes. To increase the beam intensity for radioisotope production and to improve the beam quality of the 66 MeV proton beam, a double-gap horizontal half-wave flat-topping resonator has been developed for the separated-sector cyclotron to operate at the associated fixed frequency. This type of flattopping resonator has never before been implemented in a cyclotron and this study is the first to show that it can be done, featuring a special characteristic not offered by other types. The resonator is reviewed against the other types of resonators that are already in use at other institutes around the world. The flat-topping voltage of the selected type has a sinusoidal half-wave distribution along a radial line in each of its acceleration gaps with the nodal points located on the injection and extraction orbits. The flat-topping voltage therefore progressively increases from zero at both of the two most critical orbits in a cyclotron to a maximum at about halfway between them. As a result will this resonator, apart from its basic function to reduce the energy spread in the beam, not decrease the orbit separation at the injection and extraction orbits in the cyclotron, as is the case with other types of flat-topping resonators. This advantageous feature implies that the beam pattern in the cyclotron is not affected in the regions of the delicate injection and extraction components and therefore will these components or the operational control of the beam not require any modification to accommodate the resonator. In order to design a resonator that will meet our requirements, the theory of the beam dynamics and resonator characteristics were studied to ascertain the expected improvement in beam quality and beam intensity when a flat-topping resonator is implemented with the cyclotron. All resonator types were considered and studied in terms of their power dissipation, voltage distribution, harmonic number, space requirements and influence on the beam. The horizontal half-wave resonator type, with two acceleration gaps, was selected as the most suitable for our application, because of its preferred electromagnetic characteristics and its geometric shape that permits the installation inside an existing vacuum chamber through an existing flange. Initially a half-scale resonator model was build to test the feasibility of such a resonator and also to verify the calculation methods. Transmission line methods and numerical field analysis in 3D were applied to determine the resonator characteristics. In the former method a computer program, POISSON, was used to calculate curvilinear squares on sections through a triangular-shaped transmission line and in the latter method a commercial computer program, SOPRANO, was used. SOPRANO is part of an internationally acclaimed suite of programs and the acquired knowledge and skill to use this state-of-the-art software for the studying and designing of such and other electromagnetic devices also put the institute amongst the front-runners in the world. The calculated characteristics of the half- and full-scale resonator models, the study of the different electromagnetic modes that resonate in close proximity to the required frequency, the heat transport modelling and the theory and implementation of the coupling and tuning devices are all in good agreement with their respective measured results and are reported in this document. This study lead the way to have the first-ever double-gap horizontal half-wave flat-topping resonator in a cyclotron successfully commissioned at iThemba LABS and the first tests with beam report very stable operation. Accelerator physicists now have another option to utilize for the establishing of flat-topped acceleration voltages.

Ce travail répond à la question de l’identification de l'équation du mouvement à partir des mesures expérimentales. Les structures considérées ont soit une soit deux dimensions. La méthode présentée utilise les méthodes inverses locales qui se basent sur les mesures du champs vibratoire stationnaire. Ces méthodes sont indépendantes des conditions aux limites qui sont inconnues pour l'observateur. Deux méthodes de sélection des modèles sont utilisées pour choisir l'équation du mouvement la plus adaptée parmi un ensemble des modèles a priori. La méthode est appliquée a des nombreux cas expérimentaux. Trois problématiques sont traités: identification de la force axiale dans les poutres et membranes, identification de l'orthotropie de la plaque et identification d'un panel sandwich épais. / This works deals with a question of identification of the equation of motion based on experimental measurements. The considered structures are either one or two-dimensional plane structures. The developed methodology employs local inverse methods based on local steady-state vibration field and it is therefore independent of boundary conditions. Two different model selection techniques are used to select the most adapted equation of motion from a set of apriori candidate models. The method is applied to various experimental case studies as identification of axial force in beams and membranes, identification of plate orthotropy and identification of thick sandwich panel model.

Ce travail de thèse porte sur les instabilités dans les structures minces hyperélastiques. Nous analysons les mécanismes de sélection du motif de flambement dans un solide prismatique fortement pré-contraint. Pour ce système, le modèle de poutre classique d'Euler-Bernoulli n'est pas pertinent du fait de cette forte précontrainte et il est nécessaire de recourir à une description 3-d pour expliquer l'apparition de modes instables de petite longueur d'onde. Notre analyse, fondée sur la théorie de l'élasticité finie incrémentale, montre que la longueur typique du motif de flambement est sélectionnée par l'importance relative de la pré-contrainte et du rapport d'aspect du solide prismatique. Nous nous interrogeons ainsi sur les limites du modèle classique et proposons une piste de réflexion pour la construction de nouveaux modèles. Celle-ci repose sur un développement à deux échelles combiné aux équations d'équilibre du système formulées sous forme faible. Il est alors facile de résoudre les équations exactes à chaque ordre, ce qui donne accès à la cinématique complète du système. Nous mettons en œuvre cette méthode pour l'exemple simple d'un barreau homogène en compression, ce qui nous permet d'établir le modèle classique d'Euler-Bernoulli à partir des équations de l'élasticité 3-d. La dernière partie de ce travail de thèse porte sur l'étude du flambement de systèmes invariants d'échelle : l'étude expérimentale et numérique de la compression d'un prisme à base triangulaire met en lumière une transition inédite d'un mode de flambement étendu vers des motifs localisés. / This Ph.D. work deals with buckling instabilities arising in thin hyper-elastic structures. We focus on instabilities arising in a prismatic solid submitted to finite incompatible pre-strains. We observe that the traditional 1-d Euler-Bernoulli beam model is not applicable to such a system because of the finite inhomogeneous pre-stress. The latter triggers short-wavelength instabilities that are note described by the classical 1-d models. We rely on the 3-d elasticity theory and propose a quantitative criterion on the ratio between the pre-stress and the cross-sectional aspect-ratio of the prismatic solid, that predicts the typical wavelength of the buckling pattern. This work questions the validity of classical 1-d models and suggests that extensions of these models are possible. We propose a method for the systematic derivation of reduced models. It relies on asymptotic expansions of the variational formulation of the equilibrium equations, starting from the complete expression of the energy. In this framework, kinematics can be entirely determined by solving the exact equations, order by order. We successfully apply this method to a homogeneous and isotropic beam submitted to a homogeneous compression and recover the classical Euler-Bernoulli beam model. In a last part of the manuscript we investigate the transition from extended wrinkling to localized creasing in a scale-invariant system made of a prismatic solid with a triangular cross-section, both experimentally and numerically.

Les technologies de fabrication additive(FA) basées sur la technique de fusion laser sur lit de poudres, telles que les procédés de fusion sélective laser (Selective Laser Melting ‘SLM’) et de fusion par faisceau d'électrons (Electron Beam Melting ‘EBM’), ne cessent de se développer afin de produire des pièces fonctionnelles principalement dans les domaines aérospatial et médical. Le procédé de fabrication additive offre de nombreux avantages, tels que la liberté de conception, la réduction des étapes de fabrication, la réduction de la matière utilisée, et la réduction de l'empreinte carbone lors de la fabrication d'un composant. Néanmoins, les pièces obtenues nécessitent une opération d’usinage de finition afin de satisfaire les tolérances dimensionnelles et l’état de surface. / Additive Manufacturing (AM) techniques based on powder bed fusion like Selective Laser Melting(SLM) and Electron Beam Melting processes(EBM) are being developed to make fully functional parts mainly in aerospace and medical sectors. There are several advantages of using AM processes like design freedom, reduced process steps, minimal material usage and reduced carbon footprint while producing a component. Nevertheless, the parts are built with near net shape and then finish machined to meet the demands of surface quality and dimensional tolerance.

Amostras de Teflón&#174 FEP e Mylar C foram submetidas a um bombardeio eletrônico por longos períodos a fim de se verificar a existência de uma possível corrente de condução na fase final de um carregamento por feixe eletrônico. Em vez disso, descobriu-se que a energia associada ao segundo ponto de cruzamento da curva de emissão eletrônica do material bombardeado varia lentamente com o tempo de exposição à irradiação. Por outro lado foram descobertos também fortes indícios de que o centróide de carga sofre um deslocamento enquanto a amostra está sendo irradiada. A componente da corrente através da amostra, associada a qualquer um desses efeitos se superpõe à corrente de condução (se existir) e acaba tornando inviável a sua observação, enquanto um ou ambos os efeitos persistirem. Na realidade não é só a energia do segundo ponto de cruzamento que varia; toda a curva de emissão característica do material, que é fundamental para se entender os processos e carga e descarga de amostras, sofre modificação com a irradiação prolongada. Além disto, dois novos métodos para se carregar uma amostra de polímero estão sendo propostos. Um deles permite carregar uma amostra positivamente, por etapas, a tensões mais elevadas que o método convencional. O outro possibilita carregar negativamente uma amostra, lançando mão do mecanismo de auto-regulação para interrupção do processo de carga, o que só havia sido feito até o momento para um carregamento positivo. Um novo método para se descarregar uma amostra usando o próprio feixe eletrônico também é apresentado. / Teflon&#174 FEP and Mylar C samples were submitted to an electron beam during long periods of time in order to examine the possible existence of conduction current in the final stages of the charging process. It was found that the energy associated with the second crossover point in the electronic emission curve of the irradiated material varied slightly with the time of irradiation. On the other hand, strong evidence emerged that the charge centroid is shifted while the sample is being irradiated. The component of the current through the sample which is associated with any of these effects is superimposed to the conduction current (if present), hampering the identification of a conduction current. In fact, it is not only the energy of the second crossover point that varies, for the whole emission curve is modified upon prolonged irradiation. This emission curve is fundamental for understanding the charging and discharging processes in the samples. In addition, two new methods for charging a sample are being proposed. The first allows one to charge the sample positively to surface potentials that are higher than those obtained in the conventional method. The other method permits the sample to be charged negatively using the auto-regulation mechanism for interrupting the charging process; this had previously been done only for charging samples positively. A new method for discharging a sample using the electron beam is also presented.

Os aços maraging pertencem a uma classe de aços de ultra-alta-resistência que diferem da maioria dos aços pelo mecanismo de endurecimento. Enquanto nos aços convencionais o endurecimento é obtido por transformação martensítica, nos aços maraging o aumento das propriedades mecânicas é obtido pela precipitação de compostos intermetálicos em uma matriz martensítica de baixo teor de carbono, em temperaturas da ordem de 480°C. Para o presente trabalho, empregou-se um aço maraging 300 na forma de barras cilíndricas com diâmetro de 8mm e comprimento de 120mm, cedido na condição solubilizada pelo Departamento de Ciência e Tecnologia Aeroespacial (DCTA). Os corpos-de-prova foram soldados em cheio empregando-se um laser a fibra de 2kW e um dispositivo especialmente desenvolvido para solda orbital. As condições de envelhecimento foram inicialmente estudadas e a melhor combinação de resistência mecânica e ductilidade foi obtidas na condição de para 460°C por 3 horas, que foi então adotada para a caracterização do comportamento em fadiga. Os ensaios de fadiga foram realizados em uma máquina de flexão rotativa (R = -1), em níveis de tensão correspondentes à faixa de vida de 104 a 107 ciclos. Os corpos-de-prova soldados a laser foram submetidos a três condições de tratamento pós-soldagem: a) envelhecimento, b) solubilização + envelhecimento e c) homogeneização + solubilização + envelhecimento. Os resultados permitiram avaliar e comparar o efeito da solda a laser, seguida por tratamentos térmicos pós-soldagem, no comportamento em fadiga do aço maraging 300. Os tratamentos pós-soldagem não resultaram em diferença significativa na vida em fadiga do aço maraging 300, o fator determinante para a duração da vida em fadiga foi a quantidade de poros presente na superfície de fratura. / The maraging steels belong to a class of ultra-high-resistance steels that differ from most of steel for the hardening mechanism. While the conventional steel is obtained by hardening for martensitic transformation in maraging steel the increase of mechanical properties is obtained by the precipitation of intermetallic phases in a martensitic matrix of low carbon, at temperatures about 480°C. For the present study, we used a maraging 300 steel in form of cylindrical bars with a diameter of 8mm and length of 120mm, received as solution treated by the Department of Aerospace Science and Technology (DCTA). The test specimens were welded using a 2 kW fibre laser coupled to a special lathe in order to obtain an orbital weld line. The ageing contitions of the steel were initially investigated using base material samples. The best combination of hardness, tensile strength and ductility were achieved for 460°C for 3 hours , which was then chosen for the fatigue characterization..The fatigue tests were conducted in a rotating bending machine (R = -1). The stress levels were found for fatigue lives in the range of 104 to 107 cycles. The laser welded test pieces underwent three distinct post-weld heat treatments: a) ageing , b) solution + ageing and b) homogenization + solution + ageing .The results allowed assessing and comparing the effect of laser beam welding, followed by post-weld heat treatments, on the fatigue behavior of the maraging 300 steel. The post-weld heat treatments no results in significant difference in fatigue life of maraging 300 steel , the determinant factor for fatigue life was the amount of pores present in the fracture surface.

Artefatos ocorrem principalmente, próximos de estruturas de alta densidade como o metal, sendo que essa alteração pode interferir na interpretação das imagens e no diagnóstico. O intuito do presente estudo é avaliar o nível ósseo ao redor de dentes e implantes por meio de dois programas diferentes e a qualidade da imagem na presença de artefatos metálicos. Foram avaliadas imagens de arquivo obtidas através do TCFC Gendex(TM) GXCB500 (Gendex Dental Systems, Des Plaines, IL, USA). Os software utilizados foram o Xelis Dental (InfinnitHealthcare, South Korea) e o OsiriX MD (Pixmeo, Geneva, Switzerland) em cortes tomográficos transaxiais de 2 mm. Foram avaliadas 145 imagens de TCFC, sendo incluídos no estudo 35 exames (as idades variaram de 35 a 83 anos), representando um total de 30 Dentes e 30 Implantes, sendo que houve uma distribuição harmônica entre os grupos em relação ao gênero e idade. Três avaliadores treinados realizaram as observações das imagens separadamente em dois períodos distintos e a qualidade da imagem foi classificada em severa, forte, moderada e mínima. Na avaliação da qualidade de imagem os implantes apresentaram uma prevalência maior de artefatos severo e forte. No entanto, as avaliações dos implantes apresentaram uma maior consistência intra-observador, demonstrando, portanto, uma maior dificuldade na classificação dos dentes. Além disso, na avaliação conjunta de dentes e implantes a consistência global dos observadores aumentou. O software OsiriX apresentou desempenho superior nas medidas lineares dos dentes e implantes na avaliação intra-examinador. Embora nenhuma das correlações intraclasse avaliadas tenha sido excelente, a grande maioria foi satisfatória, com valores entre 0,4 e 0,7. Podemos concluir que a presença de artefatos metálicos limita a qualidade da imagem do nível ósseo ao redor dos dentes e implantes, assim como o programa OsiriX apresentou resultados nitidamente superiores na avaliação das medidas lineares. / Artifacts occur mainly near high-density structures, such as metal, this alteration may affect images interpretation and diagnosis. The aim of this study is to evaluate the bone level around teeth and implants through two different software and image quality in presence of metal artefacts. Archival images obtained were evaluated by CBCT Gendex(TM)GXCB500 (Gendex Dental Systems, Des Plaines, IL, USA). The software used was Xelis Dental (InfinnitHealthcare, South Korea) and OsiriX MD (Pixmeo, Geneva, Switzerland) in transaxial tomographic slices of 2 mm. We evaluated 145 CBCT images and included in the study 35 exams (ages ranged from 35 to 83 years-old), representing a total of 30 teeth and 30 implants, there was a harmonic distribution between groups in relation gender and age. Three trained examiners made observations of images separately in two distinct periods and the image quality evaluation was classified in severe, strong, moderate and minimal. Evaluation of implants image quality presented a higher prevalence of severe and strong artifacts. However, implants evaluation presented a higher intra-observer consistency, therefore teeth assessment presents a greater difficulty. In addition, joint assessment of teeth and implants increased global consistency of observers. OsiriX software outperformed intra-examiner linear measurements of teeth and implants. Although none of the intraclass correlations evaluated has been excellent, vast majority was satisfactory, with values between 0.4 and 0.7. We conclude that presence of metal artifacts limits the image quality of bone level around teeth and implants as well as software OsiriX presented significantly better results in the evaluation of linear measurements.

A deformidade dentofacial de Classe III está associada a importantes alterações funcionais, psicológicas e estéticas. O objetivo neste trabalho foi avaliar a morfologia craniofacial de indivíduos adultos, com deformidades dentofaciais de Classe III comparando-os com um grupo de indivíduos com faces equilibradas com finalidade de diagnóstico e prognóstico ortodôntico e cirúrgico. Foram avaliados 50 indivíduos, os quais foram submetidos ao exame de tomografia computadorizada de feixe cônico (i- Cat®/ Cone beam 3-D Dental Imaging System, Imaging Sciences International, Hatfield, USA) com o propósito de avaliação e diagnóstico. A amostra foi dividida em dois grupos, sendo 25 indivíduos com equilíbrio das relações faciais e oclusão dentária excelente e 25 com maloclusão Classe III sendo que estes fazem parte da clínica de preparo ortodôntico para Cirurgia Ortognática da Faculdade de Odontologia da FOUSP. Após a obtenção dos arquivos digitais foram realizadas as reconstruções das imagens em três dimensões nos softwares Dolphin 3D® (Dolphin Imaging and Management Solutions, Chatsworth, CA EUA) e Maxilim® (Medicim- Medical Image Computing, Mechelen-Bélgica). Seqüencialmente foram demarcados pontos de referência anatômica e a seguir foram mensuradas grandezas lineares angulares e proporcionais que foram representadas, após avaliação estatística, por média e desvio padrão. Para determinação do erro do método foram calculadas as correlações intraclasses para avaliação da concordância e a fórmula de Dahlberg. Para comparação entre os grupos foi aplicada a Análise de Variância (ANOVA) com dois fatores, destinada a representar a oclusão e gênero. No caso de ocorrência de interação estatisticamente significativa foram realizadas comparações múltiplas de Tukey para verificar em qual dos gêneros dentro de cada grupo ocorreu diferença. O nível de significância adotado foi o de 5%. Com relação ao erro do método foi observada alta reprodutibilidade na avaliação intraoperador nos softwares Dolphin 3D® e Maxilim®. Na comparação cefalométrica 3D entre os grupos avaliados, foram observadas diferenças significativas (p<0,05), sendo que o grupo com deformidades dentofaciais Classe III demonstrou variações para o relacionamento entre altura e largura facial, associada ao aumento da porção inferior da face; rotação horária da mandíbula relacionada à maior inclinação do corpo e do ramo mandibular e pelo predomínio da altura facial anterior em relação à posterior; diminuição do comprimento e do ângulo da base craniana; compensação na altura alveolar da maxila; predominância da dimensão mandibular na configuração ânteroposterior do esqueleto facial e tendências compensatórias verticais que reduziram ao nível alveolar, os desequilíbrios esqueléticos estruturais. / Class III dentofacial deformity is associated with important functional, psychologic and aesthetics changes. The objective of this study was to evaluate the craniofacial morphology of adults, with Class III dentofacial deformity comparing them with a group of subjects with harmonious faces for diagnosis and prognosis in orthodontic and surgical cases. This study evaluated 50 subjects that were submitted to the cone beam computed tomography (i-Cat®/ Cone beam 3-D Dental Imaging System, Imaging Sciences International, Hatfield, USA) with the purpose of diagnosis. The sample was divided in two groups, one with 25 subjects with harmonious faces and excellent occlusion and another with 25 subjects with Class III malocclusion, that are in treatment in the orthodontic clinic for orthognathic surgery in the Faculty of Dentistry of University of São Paulo. After the digital archives were obtained, the 3D reconstruction of images were performed with the softwares Dolphin 3D® (Dolphin Imaging and Management Solutions, Chatsworth, CA EUA) and Maxilim® (Medicim- Medical Image Computing, Mechelen-Bélgica). After demarcation of the reference points, linear, angular and proportional measurements were made and represented in statistical analysis, by mean and standard deviation. To determine the error of method, intraclass correlations were calculated, and Dahlbergs formula was used. To compare both groups the Variance Analyses was applied (ANOVA) with two factors, representing occlusion and gender. When statistical significant differences were observed, Tukeys multiple comparison was performed to evaluate in which gender the variance happened. The level of significance adopted was 5%. The error of method showed high reproducibility for intra-operator evaluation in Dolphin 3D® and Maxilim® softwares. Cephalometric 3D evaluation of the groups, showed significant differences (p<0,05), and the group with Class III dentofacial deformity showed changes in the relationship of height and width, associated with an augmented height of the lower face; clock-wise rotation of the mandible related to an increase of the inclination of the mandible body and ramus and a greater anterior facial height in relation to the posterior facial height; decrease in the size and angulation of the cranial base; compensation in height of dentoalveolar maxilla; predominant mandibular size in anterior-posterior configuration of the facial skull and vertical dentoalveolar compensations that reduced to alveolar level the structural skeletal unbalance.

A maioria das calcificações em tecidos moles na região de cabeça e pescoço são achados radiográficos incidentais. Com o crescente uso da Tomografia Computadorizada de Feixe Cônico na Odontologia (TCFC) há um aumento do número desses achados, visto que o exame permite a visualização das estruturas em três dimensões. Esse estudo estabeleceu uma metodologia para identificar algumas dessas calcificações. Um observador calibrado analisou 100 exames de TCFC e as respectivas panorâmicas, quanto à presença de Ossificação do Complexo Estilo-Hióideo (OCEH), tonsilólitos e ateromas. Adicionalmente os processos estilóides foram mensurados. As correlações para as radiografias panorâmicas foram estatisticamente significante entre idade e comprimento do processo estilóide. As correlações para os exames de TCFC foram estatisticamente significantes entre idade e tonsilólitos, idade e o comprimento do processo estilóide e idade e ateromas. Houve diferença estatística significante (Wilcoxon p<0,05) entre os exames de TCFC e panorâmicas quanto à presença de tonsilólitos, presença de OCEH e para as mensurações dos processos estilóides. Foi detectada maior quantidade de calcificações em tecidos moles nos exames de TCFC. A identificação das calcificações em tecidos moles é importante no diagnóstico diferencial de muitas patologias incluindo os flebólitos. Portanto o profissional deve ficar atento a correta interpretação dessas estruturas, buscando evitar erros e omissões, para que possa oferecer ao paciente opção correta de tratamento se for necessário. / The most common soft tissue calcifications in head and neck region are incidental findings in radiographic images. The use of Cone Beam Computed Tomography in Dentistry, has increased these incidental findings mainly, because CBCT allows a third dimension view. The goal of this study was to differentiate the styloid chain ossification (SHCO), tonsilloliths and calcified atheromas. Based on a specific methodology, one calibrated observer analysed 100 panoramic and CBCT exams from the same patients regarding these alterations. Afterwards, the styloid process was measured at the same exams. The correlations tests for the panoramic exams were statistically significant between age and styloid process length. The correlations tests for CBCT exams were statistically significant between age and tonsilloliths, age and styloid process length and age and calcified atheromas. There was a difference statistically significant (Wilcoxon p<0.05) between CBCT and panoramic exams regarding: presence of tonsillolith, presence of SHCO and styloid process length. It was detected more quantity of soft tissues calcifications in CBCT exams. The identification of soft tissues calcifications is important for the differential diagnoses of many pathologies including phlebolits. Therefore the professional should be able to do a correct image interpretation in some cases in order to avoid mistakes and offer the patient a treatment if is necessary.

Neste trabalho apresentamos o início de operação do sistema injetor do Mícrotron do IFUSP. São apresentados estudos inéditos do tratamento de desalinhamentos tanto de uma única lente quanto de um conjunto delas. As lentes magnéticas deste estágio foram alinhadas com precisão melhor que 0,18 mm. Estabelecemos um sistema de aquisição de imagens do feixe e com ele fizemos medidas da emitância, cujo valor foi determinado em (2,32 ± 0,05) pi.mm.mrad, independentemente da tensão de aceleração no canhão de elétrons, devido à limitação imposta pelo colimador da entrada do chopper. / In this work we present the commissioning of the IFUSP Microtron injector system. We developed a new method to treat misalignments on a single lens as well as in a group of them. We installed an image acquisition system to acquire beam images from the fluorescent screen monitor. Emittance was measured and found to be (2,32 ± 0,05) pi.mm.mrad, independently of the beam energy, showing the limitation imposed by the collimator placed at the entrance of the chopper cavity.